1961~2010年大渡河流域极端降水事件变化特征

基于大渡河流域1961~2010年逐日降水数据资料,运用Mann-Kendall非参数检验、Morlet小波分析法,分析了近50a来大渡河流域极端降水事件的时空变化特征。结果表明,大渡河流域的极端降水指数均呈现出相对稳定的波动增加;多年平均值均呈现出由西北向东南方向逐渐增多的分布特征,变化趋势的空间分布存在着区域差异:除强降水日数外,其他极端降水指数均呈现下游增加,上游减小的变化趋势,大渡河流域极端降水与年降水量变化趋势密切相关。大渡河流域各指数突变特征不一致,1d、5d最大降水量突变年集中在1974~1976年前后;强降水日数、极端降水量及极端强降水日数发生突变的年份分别为1984年、1979年及1977年,且突变后呈现明显的增大趋势。大渡河流域极端降水指数周期特征较复杂,但普遍存在5~10a的年际振荡周期和20~25a的年代际振荡周期,且25a是最强的主周期。

CLIMATIC CHARACTERISTICS OF EXTREME PRECIPITATION EVENTS IN THE DADU RIVER BASIN DURING 1961-2010

In recent years, with the increase of extreme precipitation events in high-latitude Northern Hemisphere land, the average intensity and frequency of extreme precipitation and in China are also on the rise. Dadu River is the largest tributary of the Minjiang River. Because of their topography, lithology, structure, slope structure, avalanches, landslides and other disasters occur frequently. Variation in precipitation extremes in the Dadu River basin from 1960-2009 were analyzed based on daily precipitation data from 17 meteorological stations within and adjacent to the Basin. This paper used a recently developed interpolator, the greens line, based on the application of green's functions of the biharmonic operator.The interpolated surface is constructed by globally summing the weighted Green's functions centered at every data point. Moreover, the methods of linear regression, Mann-Kendall test and continuous wavelet transform were employed to delineate the rate of change, abrupt change points, statistical significance of the trends, and periodicity of extreme precipitation indices. The results showed inter-annual variations of the extreme precipitation index in Dadu River showed relatively stable volatility increases. The distribution of annual average showed an increased trend from the northwest to southeast, and there were significant regional differences in the trend of extreme precipitation index. Extreme precipitation indices showed a trend of reducing in upstream and increasing in downstream, with exception of the heavy precipitation days. The extreme precipitation indices were closely associated with annual total precipitation in the basin. The abrupt change point of each extreme precipitation index in the Dadu River basin was inconsistent. Abrupt change points of RX1 day, RX5 day were at 1974-1976, and 1978, and R25, R95 and R95 day were at 1984, 1979, 1977, and increased after abrupt changes. The cyclying characteristics of extreme precipitation indexes is more complex, we conclude that most extreme precipitation indices in the Dadu River basin existed two common oscillation cycles, 5-10 years and 20-25 years(major cycle).